I am participating in a design competition based on lunar In-Situ Resource Utilization (ISRU). A key component of this design is transporting regolith-mined water from a water extractor in a Permanently-Shaded Region (PSR) to a delivery site. The delivery site is approximately 4 km horizontal distance and 450 m above the water extractor.

Ground temperature inside this PSR is between 40 and 100 Kelvin, with the possibility of using some heating power to warm components.

A thrower-catcher system has by far the lowest power consumption of all considered water transport options, except for maybe an (extremely heavy) 4.5 km rigid pipe. The water would be in homogenous ice form, probably a cube or sphere for handling simplicity.

From kinematics, initial velocity would need to be 174 m/s at a minimum. That speed would require a huge trebuchet, but motors can be used to augment a much smaller trebuchet ("flywheel trebuchet").

Has anyone calculated the possible accuracy of using a space-grade trebuchet in vacuum? How about other thrower options such as a ballista, compressed gas cannon, chemical propellant cannon, or railgun?

Other info:

  1. Ice sublimates in vacuum, so transport either needs to be quick (<2 minutes?) or include some sort of container/wrapper around the ice.
  2. Let’s go ahead and set a 3000 kg mass limit for all components besides regolith and water ice.
  3. A 20x20 meter target (net?) is somewhat feasible to construct.
  4. Wasting ice is tolerable (up to ~90% wasted).

FYI for people who find this answer while researching for the same competition:

You must comply with this site's rules for redistribution and clearly indicate use of information from this question or answers in the contest submission. (see also).

  • 1
    $\begingroup$ Interesting. Even if the launch is feasible, the landing may shatter your giant snowballs. Could you simply roll your giant snowballs to their destination? $\endgroup$
    – DrSheldon
    Commented Mar 19, 2021 at 11:51
  • $\begingroup$ Depending on the catching hardware, that might be ok! $\endgroup$ Commented Mar 19, 2021 at 11:52
  • $\begingroup$ Why are you not distinguishing between water and ice, please? $\endgroup$ Commented Mar 20, 2021 at 18:37
  • $\begingroup$ @RobbieGoodwin it doesn’t matter how the H20 is transported, as long as it arrives at the delivery site. The output of the water extractor is liquid water, which is trivial to freeze into any shape needed. $\endgroup$ Commented Mar 20, 2021 at 19:58
  • $\begingroup$ @CourageousPotato Sorry you didn't realise this, and whether H2O is transported as water or ice is vital. The mass remains the same; both the volume and the maneuverability change hugely. You seem to be ignoring the actual Question, which dealt only with trebuchet-based materials transport… how could that apply to liquid water? $\endgroup$ Commented Mar 20, 2021 at 20:19

4 Answers 4


It’s an interesting idea, but the problem will be accuracy. Depending on the conditions found on the ground, it might be difficult to produce uniform “ammunition” and an “engine” with a sufficiently reproducible throw. The catcher would also have to be a substantial structure capable of dealing with significant wear.

I would use a cable car / chairlift arrangement. On Earth support pylons can have very large spans and in 1/6 g on the Lunar surface the span possible should be truly massive.

Cable car pylon span

Alternatively if you are determined to use a throwing engine, it might be possible to throw chunks of ice to a sufficiently shaded smallish crater or depression that was much closer to the target location. Accuracy would be less of a problem and ice debris should collect at the bottom of the crater from where they could be transported out via a much shorter pipe or by other means.

  • 3
    $\begingroup$ Great point. Probably even simpler than my slide idea and simplifies use of airtight containers to prevent sublimation $\endgroup$
    – Dragongeek
    Commented Mar 19, 2021 at 16:45
  • $\begingroup$ Alternate terminology for such systems are: aerial ropeway, ropeway conveyor or aerial tramway, material ropeway $\endgroup$
    – Fred
    Commented Mar 20, 2021 at 7:34
  • 1
    $\begingroup$ How brittle would cables get in eternal darkness? $\endgroup$
    – Jens
    Commented Mar 20, 2021 at 10:11
  • $\begingroup$ @jens stainless steel is routinely used for liquefied natural gas piping in the same temperature range as the lunar night - it's fine for the temperature. It's also used by Spacex in their starship for the same reason. The consequences of a cable faiiling on the Moon would be much less than the consequences of a liquid methane tank failng on Earth, so I think it'll be fine in cables for this application. Ideally you want an L grade (low carbon) - such as 304L or 316L for reduced brittleness. $\endgroup$ Commented Mar 21, 2021 at 10:16
  • 2
    $\begingroup$ @LevelRiverSt But cables are not pipes. The cables carry the gondolas and move. The cables undergo a U-turn at the terminals (radius a few meters), thus must be flexible. Can stainless steel do that at, say, 100K? Day in, day out? $\endgroup$
    – Jens
    Commented Mar 21, 2021 at 12:13

I'm not sure if a thrower-catcher system is a good solution. Some points:

  • Getting a canon (or launcher of any sort) to fire accurately across four kilometers without precision machined shots or a guidance system is going to be very, very difficult. Especially considering the lack of atmosphere which makes aerodynamic flight stability and control infeasible, I don't think it would be possible to hit a reasonably sized collection net. Even a high-end military sniper rifle which is firing high tolerance bullets through a precision machined barrel can't shoot four kilometers accurately, and you're throwing unbalanced chunks of ice.
  • Immense safety hazards. Even if it's all unmanned, a chunk of ice heavy enough to be useful which is also flying fast enough to travel four kilometers is dangerous enough to punch holes through pressure hulls, shatter solar panels, and cause all sorts of problems. For example, let's say a chunk misses and hits the regolith, scattering fine, ultra-abrasive dust everywhere. This dust then settles on machinery, solar panels, camera lenses, and gets everywhere causing even more issues.

Here's what I'd do:

Lunar regolith transport system image

Simply elevate the ice/regolith to a large height and then let it slide down a slide. If the angle of the slide needs to be steeper, simply use more towers. This way approaches the energy efficiency of a launching system since you're only expending energy lifting the material upwards and can simultaneously be very lightweight because the long slides can just be fabric/flexible. On Earth, a self-stacking construction crane can reach about 80 meters, so in the reduced lunar gravity, you could stack a lightweight truss tower rather high. Then, simply add a vertical conveyor belt and a slippery-but-durable material for the slide, and you're done. With a system like this you can even pass hardwired data or power through the slides, and if you need to decrease friction on them, you could even slightly heat them to melt the ice a tiny bit. Hell, due to the flexible nature of the slides, you could even have one end of this system be mobile and simply adjusting the amount of slide-material it has unspooled to the nearest tower.

  • 7
    $\begingroup$ Rifles have most of their inaccuracy driven by wind and other atmospheric conditions. I did some back-of-the-napkin math and came up with a 8.6 MOA radius to hit a target within a 10 meter radius at 4 km. $\endgroup$ Commented Mar 19, 2021 at 11:41
  • 1
    $\begingroup$ or, if it's sturdy enough, you can make the slide vibrate. Serves both to reduce friction and, as a side bonus, as a little bit of pulverization as well $\endgroup$ Commented Mar 19, 2021 at 18:18
  • 7
    $\begingroup$ Intuitively, it's not clear to my why accuracy would be a problem. So long as the mass of the projectile is known with precisions, it seems plausible to launch it through vacuum accurately. A military sniper rifle misses at 4km on Earth because of the atmosphere. $\endgroup$
    – Harabeck
    Commented Mar 19, 2021 at 19:11
  • 1
    $\begingroup$ Well, counterweight trebuchets are very inaccurate. You would need something with a barrel. Even then, I doubt you can form an ice projectile so that it would have a perfectly homogeneous density and friction all around. If you're using a gas gun and one side of the projectile has a leak or is more friction-y against the barrel, that tiny different when leaving the barrel could result in a very large error over 4 km. It's definitely doable, but in the end you'll spend too much effort carving and machining the ice chunks to be good projectiles. $\endgroup$
    – Dragongeek
    Commented Mar 19, 2021 at 19:59
  • 1
    $\begingroup$ What makes a slide better than a pipe? $\endgroup$ Commented Mar 19, 2021 at 23:30

You could build a much safer, if less dramatic, system by transporting slugs of ice through 4.5 km of thin-wall flexible tubing in the manner of a pneumatic mail tube. If you have water ice, both hydrogen and oxygen are available through electrolysis to use as the working gas. Use whichever gas works better. If you can build such a system using two tubes -- one to transport the ice and a thinner one to return the gas to the origin -- at a mass of 0.67 kg/m, the whole thing would mass no more than the 3000 kg that you allow. If left to run continuously with a slug leaving every few seconds, you could move a great deal of ice. If 4.5 km is too far for pneumatic transport, use booster stations at the cost of additional mass.

  • $\begingroup$ That sounds heavy. $\endgroup$
    – ikrase
    Commented Mar 20, 2021 at 11:20

This answer is more a comment based on another answer (but too long to fit within a comment).

If you're dead set on using a trebuchet type launcher, then the problem (just as everyone else has identified) will be accuracy. Here's a mixed system - as suggested by Slarty a cable car type set up would handle the accuracy 100%, but if you build the cable car system in such a way that it follows a roughly parabolic trajectory, there's no reason (that I can see) that you can throw the ice while it's attached the the cable car system. I just tried it with a washing line and a toilet roll, I literally threw the toilet roll along the washing line. Worked OK-ish.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.